Jason Zweig, writing in the Wall Street Journal, recently published a well-balanced article on low-volatility investing (The High Price of 'Low Volatility' Funds).
The article mainly serves as a cautionary tale for investing in the latest fads; in this case, the popularity of the low-volatility style of investing given its strong risk-adjusted outperformance over broad equities. However, the piece endeavors to represent the views of low-volatility advocates and their rationales for why this strategy makes sense.
Zweig's primary concern is similar to the one voiced by Research Affiliates: Low-volatility investing's popularity has pushed its historical valuation to unprecedented levels relative to both its own history and to the broader markets, leaving investors vulnerable to future disappointments.
Valuation Can Be A Poor Predictor
Indeed, Research Affiliates (RA) warned of the dangers of low-volatility investing as far back as the third quarter of 2013. Using valuation data through May 2013, RA contended that "low-vol strategies" had been trading at valuations well above their prior 10-year history as well as above the full sample period.
However, despite the danger signs flagged by RA, low-volatility investing continued to deliver competitive risk-adjusted returns versus the broader market (Table 1).
And despite trading at historically high valuations, low-volatility strategies delivered on their primary desirable attribute; namely, lower volatility (standard deviation) versus the broader markets.
Table 1: Low-Volatility Investing Delivers on Lower Volatility
*Data source: Bloomberg. Annualized returns and standard deviations based on monthly returns of indices representing styles covering market capitalization-weighting, low-volatility weighting (based on MSCI and S&P's methodologies) and fundamental weighting (FTSE RAFI Indices). For a larger view, please click on the image above.
In response to publications calling into question the "valuations" of alternative beta strategies, such as low-volatility investing, Cliff Asness of AQR Capital Management published an editorial piece in a forthcoming issue of the Journal of Portfolio Management titled "The Siren Song of Factor Timing."
The article questions the forecasting ability of valuation-based spreads (e.g., book-to-price spreads between the highest-ranked versus lowest-ranked securities) within a factor to determine future factor profitability.
Valuation only seems to work at extreme moments—such as the 1999/2000 Internet bubble or the aftermath of the 2008 global financial crisis—but factor timing also presumes that, by using valuation spreads, one can consistently predict when such extreme scenarios will materialize.
Of course, trying to time these types of crashes is as difficult as trying to time a market crash itself. Larry Swedroe on ETF.com also summarized recent academic work questioning the use of valuation as a factor-timing tool. Asness does not rule out the possibility of formulating a factor-timing strategy, and based on my own experience of testing quantitative factor timing strategies, I believe there is the potential to incorporate other metrics in factor timing such as trend-following rules and warnings signs indicating overcrowdedness, such as short interest levels.
However, even if a holy grail of factor timing could be uncovered, Asness makes this astute point:
" … unless we one day see these [factor] strategies also richen to unprecedented unsupportable levels, not something we see today, I see this as an unfortunate pain we must bear in exchange for the long-term positives of good factors. Again I make the analogy to knowing that the stock market will one day suffer a short painful 'crash' does not mean one doesn't invest in stocks for the long run."
Know What You're Buying
Despite Zweig's warning with respect to the valuation premium that low-volatility funds are trading at versus the broader markets (when historically they've traded at a discount), investors view low-volatility style as one of the better ways of maintaining market exposure without engaging in market timing.
The two most popular low-volatility ETFs are the iShares MSCI USA Minimum Volatility (USMV | A-69) and the PowerShares S&P 500 Low Volatility (SPLV | A-58), although some of the recent proliferation of multifactor ETFs such as the Goldman Sachs ActiveBeta U.S. Large Cap Equity (GSLC) incorporate a low-volatility factor.
Table 2 displays a characteristic breakdown between USMV, SPLV and the S&P 500 Index, as proxied by the SPDR S&P 500 ETF Trust (SPY | A-97). Table 3 displays the relative sector weighting differences of USMV and SPLV versus the S&P 500. All characteristic data comes from Bloomberg.
Table 2: Fund Characteristics
Table 3: Relative Sector Weights (vs. S&P 500 Index)
Based on the fund characteristics, we can observe the following:
- Low volatility trades at a premium to the S&P 500 on a forward-earnings and book-value basis.
- Low volatility also has a higher projected dividend yield, suggesting it is more susceptible to interest-rate movements (as implied in the WSJ article), although not as sensitive as pure dividend-focused ETFs.
- Based on sector deviations, one can argue that SPLV represents a purer play on low-volatility than USMV, since the latter constrains its relative sector deviations to 5%. However, with higher exposure to low volatility comes greater sector risk, as SPLV is much more concentrated in consumer staples and utilities versus USMV.
Running USMV and SPLV through Bloomberg's risk model, we find the following:
- USMV has an equity beta of 0.80, while SPLV has an equity beta of 0.73. These are consistent with what you would expect from this portfolio, with SPLV having lower risk than USMV, given the former's construction methodology.
- USMV has a forecasted tracking error (or active risk) of 4.6% versus the S&P (meaning that USMV is expected to perform within +/- 4.6% versus the S&P over a majority of one-year periods), while SPLV has a projected tracking error of 6.0%. Again, this is consistent with SPLV's construction methodology allowing for greater sector deviations.
- Breaking down the sources of active risk, the majority come from industry risks, as would be expected, given the large relative sector weights.
- However, the factor exposure breakdown gives a more interesting story. Apart from "size" or smaller-cap risk exposure, Bloomberg's risk measure of low volatility is the primary contributor to SPLV's overall risk-factor exposure, followed by momentum, earnings variability and growth. For USMV, momentum serves as the primary contributor, followed by low volatility and value.
By conducting an X-ray on low-volatility ETFs, investors can get a more granular picture on what it is they are buying. SPLV represents a purer play on low-volatility investing, while USMV represents a meaningful play on low volatility, but tempers this exposure through commonsense guardrails such as sector constraints.
Don't Bet Only On Low-Volatility Investing
Personally, if I had to bet on what strategy will perform 12 months out, the contrarian in me leans toward value investing over momentum and low volatility, given the popularity of the latter two.
Indeed, we've seen a bit of this reversion play out in March and April. But the near-term matters little when incorporating alternative betas into a broad equity strategy designed to capture long-term risk premia.
Academics struggle with the low-volatility anomaly because it turns capital asset pricing models on their heads (investors should be compensated with excess premia for taking on higher risk, not the other way around).
In an environment of low anemic growth and weak inflation, low-volatility investing remains the preferred method of gaining market exposure. But just as Cliff Asness advised, one should not bet solely on a single factor or strategy, but incorporate it as part of a diversified portfolio.
At the time of this writing, 3D Asset Management held SPLV and USMV. The above is the opinion of the author and should not be relied upon as investment advice or a forecast of the future. The projections or other information regarding the likelihood of various investment outcomes are hypothetical in nature, do not reflect actual investment results, and are not guarantees of future results. It is not a recommendation, offer or solicitation to buy or sell any securities or implement any investment strategy. It is for informational purposes only. The above statistics, data, anecdotes and opinions of others are assumed to be true and accurate; however, 3D Asset Management does not warrant the accuracy of any of these. There is also no assurance that any of the above is all-inclusive or complete. Past performance is no guarantee of future results. None of the services offered by 3D Asset Management are insured by the FDIC, and the reader is reminded that all investments contain risk. The opinions offered above are as of May 9, 2016, and are subject to change as influencing factors change. More detail regarding 3D Asset Management, its products, services, personnel, fees and investment methodologies are available in the firm's Form ADV Part 2, which is available upon request by calling (860) 291-1998, option 2, or emailing [email protected] or visiting 3D's website at www.3dadvisor.com.